This is the mail archive of the cygwin@sourceware.cygnus.com mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Helping out on the list


At 10:16 24/9/97 -0400, Chris Faylor wrote:
>In article <memo.19970923183002.36397A@horus.cix.co.uk> you write:
>>Asking for help isn't one of my string points, I'll grant; I prefer to try
>>and suss things out for myself.
>
>Hi,
>That's the right way to do it!  I'm actually getting pretty sick of
>all of the posts to the mailing list that are essentially:
>
>"Hi.  I am a newbie and don't know how to use your software.  Apparently
>I have to 'down' 'load' it or something but I can't find any documentation
>immediately in front of me on the screen so I thought I'd ask for help
>because I *do* know how to type."
>
>I think that trying to figure things out for oneself might be a dying
>discipline.  It seems that it may be supplanted by blaming the people
>who provided you with whatever you're trying to use of not being able
>to adequately explain things to you.
>
>Sorry about this rant.  I really just wanted to let you know that I
>admire people who try to "suss things out" themselves.  Keep it up! It
>probably won't be a trend but in a twenty years or so you might actually
>be worshipped for your supernatural ability to figure things out without
>begging for help.  Either that or you (and I) will be burned at the
>stake...

Ah, thirty years ago, you would have been saying: "Let 'em reverse engineer
the binaries if they want documentation. That's the way I did it!" :-)

I agree that some people ask questions on a list when the answer is right
there in front of them. The proper response is: "The answer to this
question is at http://blah.com/foo.html#the.answer."

But think again: How did *you* learn most of this stuff? By puzzling it
out, by asking someone, and maybe even reading the doc.

The problem thirty years ago was a dearth of doc. We were too busy writing
the software to bother writing the doc. Now, the problem is that we have
too much doc, and it's badly written and poorly organized. How many
megabytes come on each MSDN CD-ROM -- and if you subscribe you get one
quarterly! How many megabytes of man pages are there out there -- but you
have to know the name of what you are looking for to use them. How often is
the typical newbie going to say, "Gee, I need to find a string in these
files. I'll try 'man grep'!"? Real obvious naming conventions here in
computer land!

I just started in on the info pages for BASH. Murphy, that's badly
organized. I couldn't care less which features came from the Bourne shell
and which came from csh. But either I have to know that to use the doc, or
I have to search in all three places to find a feature. And I still haven't
found what I am looking for: the name of the auto-initialization script for
BASH, and where to put it ($HOME, I presume). Later today, I'll fire up
Linux and read the man page to find it.

Every one of us on this list is here because we are using "free" software
-- GCC, BASH, or cygwin32.dll. We owe the people who wrote those tools for
the hundreds of hours we did not have to put in writing those tools
ourselves (and I know -- I've written compilers, assemblers and OSs). How
do we pay them back for doing all that work?

The answer is, we don't. We pay them forward. We help out the next guy.

You are right that figuring things out for yourself is a dying discipline.
Possibly because people are getting lazy, or have gone to government
schools, but I doubt that that explains it all. I think a better explantion
is that software is much more complicated than it was even ten years ago --
and therefore harder to figure out.

I know a guy who invented a programming language, and wrote an OS in it. He
and his colleagues ported it to many different processors and environments.
Later, he designed his own processor to execute that language, and did it
on chip layout software he wrote in his own language. He sold the processor
along with a kit. The kit included a pc board which he laid out on a
package he wrote for the purpose. Try *that* with Windows or any modern Unix.

The language is simple enough that anyone of reasonable intelligence can
understand it well enough to port it, and write the tools to port it. I
know -- I ported it to five different processors, and a customer of mine
ported it to another one. I did three versions for the 68000 alone. Try
that with Windows or any modern Unix.

Of course people are asking more questions. Software is far more complicated.

Don't get me wrong: I am fully in favor of people working it out for
themselves. But don't expect them to find all the answers. And help them
out when they ask for it. Remember, someone may just help you out when you
ask.

Of course, we all once in a while ask a question which is documented. I
just asked how to add the Cygnus info files to my emacs info tree, and got
three answers on the list. The answer is right there in the doc in Emacs'
Info tree. Oops. And thanks for the answers, people.


		-- C^2

Looking for fine software and/or web pages?
http://web.idirect.com/~ccurley
-
For help on using this list (especially unsubscribing), send a message to
"gnu-win32-request@cygnus.com" with one line of text: "help".


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]